The soil and groundwater of former ordnance plants and their dumping sites have often been highly contaminated with the explosive 2,4,6-trinitrotoluene (2,4,6-TNT) leading to a potential hazard for humans and the environment. Further hazards can arise from metabolites of transformation, by-products of the manufacturing process, or incomplete combustion. This work examines the toxicity of polar nitro compounds relative to their parent compound 2,4,6-TNT using four different ecotoxicological bioassays (algae growth inhibition test, daphnids immobilization test, luminescence inhibition test, and cell growth inhibition test), three genotoxicological assays (umu test, NM2009 test, and SOS Chromotest), and the Ames fluctuation test for detection of mutagenicity. For this study, substances typical for certain steps of degradation/transformation of 2,4,6-TNT were chosen for investigation. This work determines that the parent compounds 2,4,6-TNT and 1,3,5-trinitrobenzene are the most toxic substances followed by 3,5-dinitrophenol, 3,5-dinitroaniline and 4-amino-2-nitrotoluene. Less toxic are the direct degradation products of 2,4,6-TNT like 2,4-dinitrotoluene, 2,6-dinitrotoluene, 2-amino-4,6-dinitrotoluene, and 4-amino-2,6-dinitrotoluene. A weak toxic potential was observed for 2,4,6-trinitrobenzoic acid, 2,4-diamino-6-nitrotoluene, 2,4-dinitrotoluene-5-sulfonic acid, and 2,6-diamino-4-nitrotoluene. Octahydro-l,3,5,7-tetranitro-l,3,5,7-tetrazocine and hexahydro-1,3,5-trinitro-l,3,5-triazine show no hint of acute toxicity. Based on the results of this study, we recommend expanding future monitoring programs of not only the parent substances but also potential metabolites based on conditions at the contaminated sites and to use bioassays as tools for estimating the toxicological potential directly by testing environmental samples. Site-specific protocols should be developed. If hazardous substances are found in relevant concentrations, action should be taken to prevent potential risks for humans and the environment. Analyses can then be used to prioritise reliable estimates of risk.